Announcement

Collapse
No announcement yet.

Apple M2 vs. AMD Rembrandt vs. Intel Alder Lake Linux Benchmarks

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #21
    The two Ngspice 34 benchmark appear to list the compiler options as " -O0 -fopenmp -lm ...". I'm at loss for why this is and how we can conclude anything interesting on an unoptimized benchmark. What am I missing?

    Comment


    • #22
      Originally posted by Ladis View Post

      Exactly. Air M1 was too amazing (Apple wanted to ensure people will like the switch to ARM64), so Apple had to cripple Air M2 to make consistent steps in the next models line.
      Apple crippled the M2 256GB model because Apple Silicon is expensive to manufacture.
      M2 consumes a bit more power (like newer Intel/AMD/NVidia chips), but Apple also decided for weaker/cheaper passive cooling (and there's no active cooling naturally in the Air models).
      M2 consumes more power because it's also made in 5nm and power efficiency has no place else to go but down.
      The storage problem is well known and affects only the cheapest model. Such capacity is not usable for people, who need to do a real work, so they take the 512GB+ models anyway.
      The 512GB model is an extra $200 more. That's 256GB more for $200. I can buy 2TB SSD for that much.

      Comment


      • #23
        Originally posted by stormcrow View Post

        Bull.
        There's software for the Apple silicon called Bull? It must be amazing.

        Comment


        • #24
          Originally posted by timofonic View Post
          Apple Silicon is overrated
          Until you compile a debug build of LLVM on a M1-Pro in less than half the time it takes a 8 core /16 thread Ryzen 3700x at less than 25% of the power usage ...

          It is 2x faster at only 1/4 the power. So 800% "better" than a R7 3700x ... (Kind of deserves the praise it is getting. In my book at least)

          It is a real pitty that we can not get performance/W graphs at this time.
          Last edited by Raka555; 09 August 2022, 04:17 PM.

          Comment


          • #25
            Originally posted by timofonic View Post
            Apple Silicon is overrated
            Indeed. Another eye opening testing was done by HU: https://youtu.be/FWfJq0Y4Oos
            The fact that M2 is even less efficient in some workloads using native apps means Apple Silicon is overrated AF.

            But to be honest it's nothing surprising. Jim Keller himself said ISA alone is not crucial for efficiency and performance. I do not know where this geek wet dream about ARM imaginary superiority over x86 was born. I guess a lot of people are watching such nonsense as LTT, where that ridiculous garbage video about ARM destroying x86 was released LOL

            Comment


            • #26
              Originally posted by Dukenukemx View Post
              Apple crippled the M2 256GB model because Apple Silicon is expensive to manufacture.

              M2 consumes more power because it's also made in 5nm and power efficiency has no place else to go but down.

              The 512GB model is an extra $200 more. That's 256GB more for $200. I can buy 2TB SSD for that much.
              Apple Silicon is actually cheap to make. In general processors have high fixed costs and low variable costs. That's why Apple gives the powerful base M1 even in iPad Mini (the fixed costs were paid by sold MacBooks).

              That 2TB SSD is not the same speed and number of overwriting like the one in Macs (that's doesn't mean Apple doesn't overprice a bit - the higher model, the more).

              Comment


              • #27
                Michael ................

                Originally posted by phoronix View Post
                An M2-powered Mac Mini could make for a very nice developer box if such an upgraded product product were to launch.

                Comment


                • #28
                  Again! Max Mini? Is that a thing?

                  for compelling ARM Linux developer laptops or boxes (Max Mini),

                  Edit: Michael there is one more typo but I am not going to report it.... enough...
                  Last edited by tildearrow; 09 August 2022, 04:49 PM.

                  Comment


                  • #29
                    Originally posted by drakonas777 View Post

                    Indeed. Another eye opening testing was done by HU: https://youtu.be/FWfJq0Y4Oos
                    The fact that M2 is even less efficient in some workloads using native apps means Apple Silicon is overrated AF.

                    But to be honest it's nothing surprising. Jim Keller himself said ISA alone is not crucial for efficiency and performance. I do not know where this geek wet dream about ARM imaginary superiority over x86 was born. I guess a lot of people are watching such nonsense as LTT, where that ridiculous garbage video about ARM destroying x86 was released LOL
                    Thanks for the link. I watched it.

                    I have 2 observations here:

                    1) They test a passive cooled laptop against active cooled laptops. Not really Apples to .. uhm
                    I can say the same for Michael's tests.
                    It would be more fair to compare a MacBook Pro to other active cooled laptops.

                    2) They should run these tests on battery power only. As that is where Apple silicon shines. Most of the contenders won't even finish all the tests and they will score much worse.

                    (I am not a MacOS fan, but these hardware impress me. Just waiting for Asahi to mature a bit more
                    Last edited by Raka555; 09 August 2022, 05:41 PM.

                    Comment


                    • #30
                      The Thinkpad X13s Snapdragon 8Cx gen 3 runs archlinux ( https://twitter.com/pierro_78/status...55496302268416 ) ... I would be interested to know how it compares to the Macbook Air M1 or M2 ...

                      Comment

                      Working...
                      X